If pip install -r requirements.txt fails with ResolutionImpossible, the usual cause on this project is using Python 3.13. Several ML packages are still more reliable on Python 3.12.
py -3.12 -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install --upgrade pip setuptools wheel
pip install -r requirements.txt- The project dependency set is pinned to broad, compatible ranges.
- Python 3.12 has much better support than Python 3.13 for common ML and embedding packages.
- Upgrading
pip,setuptools, andwheelfirst avoids older build backends causing false resolver failures.
Delete the old .venv folder and recreate it with Python 3.12 using the commands above.
This project is an AI Education Bot that answers questions using semantic search with Endee vector database.
- Semantic search over educational content
- Simple Flask web interface
- Highlights best answer based on query similarity
- Easily extendable with more educational content
- Python
- Flask
- Sentence Transformers
- Endee vector database
- Clone the repository
- Install dependencies: